Improve model card: Add metadata and sample usage

README.md

@@ -1,3 +1,9 @@
+---
+license: mit
+library_name: transformers
+pipeline_tag: image-text-to-text
+---
+
 🌟 **ARES** — Adaptive Multimodal Reasoning Framework  
 Two-stage adaptive reasoning: cold-start + entropy-shaped RL.  
 
@@ -10,8 +16,46 @@ SOTA efficiency–accuracy tradeoffs on diverse multimodal and textual benchmark
 2. **Entropy-Shaped RL (AEPO)** — trigger exploration via high-window entropy, hierarchical rewards  
 
 📂 Resources  
-- **Paper**: ARES: Multimodal Adaptive Reasoning via Difficulty-Aware Token-Level Entropy Shaping :contentReference[oaicite:0]{index=0}  
-- **Code**: [GitHub – shawn0728/ARES](https://github.com/shawn0728/ARES) :contentReference[oaicite:1]{index=1}  
+- **Paper**: [ARES: Multimodal Adaptive Reasoning via Difficulty-Aware Token-Level Entropy Shaping](https://huggingface.co/papers/2510.08457)  
+- **Code**: [GitHub – shawn0728/ARES](https://github.com/shawn0728/ARES)  
+
+---
+
+## Sample Usage
+
+Run the command below.
+```python
+MODEL_PATH="ARES"
+MAX_TOKENS=16384
+DO_SAMPLE=True
+TEMPERATURE=1.0
+TOP_P=0.95
+TOP_K=50
+NUM_RETURN_SEQUENCES=1
+
+
+prompt = "You FIRST think about the reasoning process as an internal monologue and then provide the final answer. The reasoning process MUST BE enclosed within <think> </think> tags. The final answer MUST BE put in \\boxed{}."
+question="xxx"
+
+
+python infer.py \
+ --model_path ${MODEL_PATH} \
+ --image_path ${IMAGE_PATH} \
+ --question ${question} \
+ --prompt ${prompt} \
+ --max_tokens ${MAX_TOKENS} \
+ --do_sample ${DO_SAMPLE} \
+ --temperature ${TEMPERATURE} \
+ --top_p ${TOP_P} \
+ --top_k ${TOP_K} \
+ --num_return_sequences ${NUM_RETURN_SEQUENCES} 
+```
+You can also modify the arguments in `inference/inference.sh`
+```shell
+bash inference/inference.sh
+```
+
+---
 
 📌 Citation  
 ```
@@ -28,4 +72,4 @@ SOTA efficiency–accuracy tradeoffs on diverse multimodal and textual benchmark
 
 ---
 
-Give **ARES** a shot and tell us what reasoning challenges it helps you solve! 🚀  
+Give **ARES** a shot and tell us what reasoning challenges it helps you solve! 🚀